evobench: Standardized Benchmarking for Evolutionary Algorithms

evobench is a Python library designed for the rigorous benchmarking of evolutionary algorithms and metaheuristics in continuous optimization.

Key Features

• Implemented Baselines Algorithms: PSO (Particle Swarm Optimization), EDA (Estimation of Distribution Algorithm), and ABC (Artificial Bee Colony).
• Benchmark Functions: A comprehensive suite including Sphere, Ackley, Rosenbrock, Schwefel 1.2, and Trid.
• Statistical Analysis: An automated decision flow that includes normality tests (Shapiro-Wilk) and comparative testing (ANOVA/Kruskal-Wallis).
• Extensible Architecture: Built upon the EvolutionaryAlgorithm abstract base class to facilitate the seamless creation of new optimizers.

Documentation Content

• Getting Started: Installation and initial setup guides.
• API Reference: Detailed documentation for classes, methods, and modules.
• Theory & Stats: Mathematical foundations and statistical methodology.
• Examples Gallery: Practical guides and implementation use cases.

Authors

Developed by Enrique Gómez Linares and Victoria Galván Delgadillo.

---

MIT License | GitHub Repository